Recent research has shown that the structure of sexual networks can increase individuals' risks for sexually transmitted diseases (SIDS) and pregnancy. Accordingly, sexual networks are currently being examined in several studies among adolescents. However, there are nonesuch studies specifically targeting Latino/a teens, even though compared to most other ethnic and racial groups, Latino teens are at increased risk for STDs and have the highest rates of pregnancy. This is a particular problem in California, and especially in the Bay Area, where Latinos represent the fastest growing ethnic group. To address this deficiency, we propose a two-year longitudinal study to examine social and sexual networks of Latino/a adolescents, 14-19 years old, from the Mission District of San Francisco. Our project has three specific aims: 1) to measure the size and structure of Latino/a adolescents' social and sexual networks, at baseline and over time, including the association between network structure and membership with reproductive health outcomes (STDs and pregnancy); 11) to identify predictors of membership in high-risk sexual networks among Latino adolescents using factors derived from the Social Cognitive Model of the Development of Ethnic Identity and from Problem Behavior Theory (such as family, peer and community influences as well as independence and desire for achievement); and III) to compare the configuration, properties, and mixing matrices of sexual and social networks of Latino/a adolescents with those of African American adolescents in a contiguous geographic area using data from a parallel, companion study. In order to characterize social and sexual networks, we propose a form of egocentric sampling based on an initial, population- based random sample of index individuals recruited without regard to infection status that then builds on snowball sampling of partners. Our cohort of initial respondents will consist of sexually active teens who have been randomly selected, one or two of their closest friends who also meet eligibility criteria, as well as randomly selected teens who are sexually inactive at baseline, but who become so over the course of the study. The network will be derived from recruitment of two generations of their sexual partners and information about their friends. Participants will be interviewed using state-of-the-art interviewing techniques (audio computer-assisted self-interview [ACASI]) and tested for gonorrhea and chlamydia (using ligase chain reaction) and pregnancy by collection of urine specimens. The study will permit detailed examination of the combined influence of epidemiologic, psychosocial, behavioral, and network factors on sexual risk taking, the acquisition of STDS, and likelihood of pregnancy.